I guess this a question about 'career pacing'. I am currently a first-year grad student studying for a PhD in Aerospace Engineering at a research university in the US. Even though I have only been studying this field for about 5 months, I am already starting to get quite a few ideas about interesting potential avenues for research. For example, possible numerical simulations or experiments that could be done.

I am thinking about pursuing an academic career post-PhD, so my question is: would it make sense for me to keep some of these ideas 'in reserve' for a possible post-doc or tenure track position? If I have one or more really 'great' ideas, when is the best time to pull those out of the bag? Now, during the PhD, or save them for later?

What I'm thinking is: if I pull out my best ideas now, it might be less clear whether they were really my ideas or my PI's; however, if I wait, then someone else might have similar ideas and beat me to it (i.e. publish or perish).

To look at it in another way: when is the optimal time during an academic career to fire off your 'biggest impact' ideas (if there is such a thing)? If I have more 'promising' ideas than I think I can work on during the PhD, would it make sense then to keep some in reserve?

  • 5
    I don't think you should. As you gain more experience and knowledge, your ideas tend to get more interesting. As an example, in the recent past, I have written many ideas down and never seem to go back to them. I always pursue my latest ideas. Jan 17, 2021 at 20:24
  • The point in paragraph 3 is valid, but maybe your current PI would be OK with you publishing as sole author. If that is the case, you would get the full credit for your ideas. Jan 17, 2021 at 20:27
  • 5
    Other people's mileage may vary, but something it took me years to learn: never hold back on an idea. The ideas you have later will be better anyway. If you try to go back to an old one you might get it finished, but working on it won't be nearly as much fun as working on the new idea that you're excited about now. If you have a good idea and you have time, then work on it now - you're not going to regret that later.
    – N. Virgo
    Jan 19, 2021 at 14:29
  • 3
    Whatever you do, don't worry about running out of ideas. They don't stop for a very, very long time. If anything, they get better as you get more experience, and they get more refined as you grow your professional network!
    – corsiKa
    Jan 20, 2021 at 22:06

8 Answers 8


I don't see a point in holding back ideas on purpose, at least not for a longer time.

If you are doing well in research, you will generally have more interesting ideas than you have time to pursue. And you will keep having new ideas all the time. So there won't be a time when you will have ample of time, but be in desperate need of a great project.

So you should work now on the best ideas you have now: First, people will judge you by your best work and best ideas. If you work on more exciting ideas now, you will be more likely to secure good postdoc positions later. And if you have a fair advisor, they will clearly credit you with those ideas (e.g. in recommendation letters). So you win by working on your best ideas now. Second, your best ideas now might be outdated in a few years time, or your interests might shift, or you might have even better and more timely ideas at that time, so there is not much point in saving those ideas. Third, someone else might have the same idea and publish a result on it in the meantime, making your ideas not new any more. And finally, even if you don't have the time to work out all ideas fully yourself, it can make sense to share them with colleagues: Maybe they get excited by the ideas and collaborate with you on them, increasing your scientific output.

Finally, what if you will not have good ideas any more in the future? Well, then very likely your chances to secure a permanent position in research are slim, and having kept your old ideas will not save you.

  • 4
    This is really helpful advice, thanks. "Working now on the best ideas I have now" makes a lot of sense.
    – Time4Tea
    Jan 18, 2021 at 1:32
  • 11
    +1 Even if you have brilliant ideas now, they most likely will be obvious to anyone in your field by the time you get your doctorate.
    – Karl
    Jan 18, 2021 at 11:07
  • I agree on following the good ideas now. However, my experience with good ideas is that the more interesting ones did not get outdated, rather, when I am asking around whether anyone has heard of a solution/tried approach X etc., the answer tends to be, no, but that sounds interesting; let me know when you have the solution :-) Jan 18, 2021 at 17:09
  • @cbeleites But aren't those rather the ideas which have always been there, yet no one knows how to actually resolve them?
    – user151413
    Jan 18, 2021 at 22:36
  • @user151413: from the feedback I got, no. Some of them turned out to be related to known problems which had totally different names in other subfields, but some of them are apparently new questions. One may say that some have become "askable" only recently, though - sometimes because other developments have paved the way. Jan 18, 2021 at 23:50

When I was a graduate student, I thought I had lots of good ideas; most of them I realized after working on them for 6-12 months were not so good or led nowhere (i.e. they weren’t good at all).

In a competitive field, there is no point in holding back. In 5 years from now the simulation you wanted to do at the start of your PhD will no longer be cutting edge or - if it’s really a good idea - someone else will have scooped you to publication.

Moreover, good ideas actually snowball. A really good idea gets noticed and opens collaborations that bring other good ideas. If the idea is yours, then these collaborators will recognize this as a fact through your comfort and competence with the and will attribute you proper credit.

  • 8
    I think the first paragraph is of major relevance. It would be the rare grad student indeed who can accurately assess the quality of their ideas four months into a PhD; learning that skill is after all one of the main goals of a PhD. Jan 19, 2021 at 18:16

I'm not sure I'd look at it quite like that. Your dissertation has a problem to solve an/or questions to answer, so you do what's necessary to resolve that. That is the main line.

There is no reason to make your dissertation longer or more complex than it needs to be.

But, along the way you will have ideas for other related (and maybe not so related) things that you might pursue. Rather than trying to find a way to squeeze them into the dissertation, creating a hodgepodge, keep a notebook of those ideas.

Make it a formal notebook, with a new idea starting on a new page, such as a physical notebook. If you use an electronic version, keep it backed up as you may not return to some of the ideas for years.

You can keep this notebook up throughout your professional life, adding to it as ideas come to you. It is also worthwhile perusing the notebook periodically to see if you have things to add to earlier thoughts. At some point some specific thread may ripen into a paper.

With regard to your concern about who's idea something is, there is no reason you can't continue a collaboration with your PI after you graduate. Collaboration at a distance is much more possible now.

  • 1
    And when the notebook is full, that's when you get tenure! :) (On a more serious note, what speaks against pursueing projects which don't make it into the dissertation?)
    – user151413
    Jan 17, 2021 at 21:02
  • 2
    @user151413, actually when all the notebook ideas get checked off is when you get tenure.
    – Buffy
    Jan 17, 2021 at 21:05
  • Thanks. I really like your suggestion of starting a 'research ideas' notebook. I will definitely do that.
    – Time4Tea
    Jan 18, 2021 at 1:36

In some fields, coming up with new ideas is not hard. But then, you still have to carry the research to test these ideas, and some of them may turn out to be bad ideas or to give results that are not great, even if you though it was some good ideas. What I personally do is to keep track of all the ideas that I have over time. For example, when I read a paper and think about some new possibility, I will take some notes to remember it. Then, when I need to start some new project or give a project to a new graduate students, I will browse through my list of previous ideas to select the best one or to use this as startup point to search for more ideas. Generally, it is better to always work on the best idea first, because you want to publish good research.

But what is the "best" idea? Well, you can think about it from different perspectives such as the potential impact in your field, the time required to do the research (some idea may be easier than others), the risk of failure, the risk that someone else publish it before you or not, etc. So I recommend to list your ideas and think about which one you should give the most priority first based on the different perspectives and your own goals. For example, if your goal is to publish a paper quickly because you urgently need to graduate, you may focus on some easier idea than working on the harder ideas. But if your goal is to get a paper in the top journals, then you should focus on the harder ideas that may have a bigger impact first.

Thus, to summarize, I think you should focus on your "best" idea first, where best is defined based on your goals and the different criteria to evaluate an idea (time, risk, potential, impact, ...) You don't need to hold back ideas but I recommend to keep a list of your ideas so you don't forget them. It may be useful later.


The answer may depend on the field you happen to be working in. As an example, in health sciences, 5 months is an extremely short amount of time. It takes years until you have any data that's worth publishing and as a junior scientist, it is rare to publish ideas or concepts alone (ie. writing a review article, letter to the editor etc.). Those articles are usually written by senior scientists in the field who are invited by the journals to do so.

The focus of a junior scientist (at least in health sciences) is expected to be the production and publishing good data. Once you have the data, the priority is to get it peer reviewed and published asap, and when you do so, you do not want to be withholding any data for later. You do not want to make a hodgepodge as Dr. Buffy said, but in order to get a high impact paper, you want your paper to be rich in content.

A paper that is rich in content gets a lot of citations, not only for the main message that it is conveying, but for the methods employed, individual pieces of data and the occasional reference to new ideas, concepts and possible future research directions (and... you are intending to withhold this?:D). A journal's aim is to maximize its impact factor, so a rich manuscript is the most valuable for them. When you try and divide your content into sub-papers, your main manuscript loses some of its desirability for the high impact journals and the sub-papers will usually be extremely difficult to publish anywhere. Some may never even make it to any journal at all, and the process will drain a lot of time and energy from you.

You also need to take into account the human factor: by the time you successfully published 1 paper, you will be near the end of the road as a PhD student, burnt-out, emotionally drained and depleted. The ideas that once looked great will appear so far away and it is not unusual for individuals to prefer spending time with family (or just some high quality sleep) over pursuing those. I found myself handing some great projects of my own to newcomers of the lab which I never regretted. Most eventually got published and I was the severalth author, but at least they made the finish line. I guess my point is, if you discuss your great ideas with the team now instead of keeping them to yourself:

  1. You may at least get some credit for the intellectual aspect of the work.
  2. The ideas may have a better chance at eventually being realized and your name being associated with it.
  3. You will get criticism, which is great. Some ideas may not be as great as you imagined they were, and may have some room for improvement.

One can't do everything on their own and 21st century science is nothing but teamwork.


Previous answers give some great advice, which is somewhat one-sided. These answers are based on the presumption of having a "fair advisor". If you were confident in that, you would have talked to them directly. But since you posted your question here, the validity of such a strong assumption is questionable.

[Speaking of one-sidedness, some participants even advocate + letting unethical behaviour stand, claiming it is for the victim's own good, which makes me suspect the class solidarity motive. It is difficult to give impartial answer to student-vs-faculty questions, because most respondents belong to one of the two groups (guess which one), but we all must at least try!]

Here are reasons why you might want to hold your idea from your PI/supervisor and some problems you may face if you don't.

Your PI may or should have plans for your PhD. Maybe they already have a great project reserved for you, and are just waiting for you to get ready for it? Sharing your idea may disincentivise your advisor from thinking hard. Pushing your idea may discourage them from caring about you. They may decide that you will be fine and it is better to "feed" other students, who are more receptive or have less on their plate. In other words, you will simply learn more by listening more and letting others speak.

Your PI's ego and favouritism. They may think it is their role to generate ideas. Or that their most senior or favourite student should do it, not you. If, in your PI's worldview, a first-year PhD student cannot propose a good idea, then it won't be a good idea or it won't be yours. No matter the reality. A spherical PI in a vacuum does not behave irrationally or immorally, but humans do.

Your PI may have plans on you and envisages you doing specific work for them. This is more relevant to postdocs, because postdocs are paid from PI's grants, which is not always the case with PhD students. Nevertheless, your PI may expect you to do the work they allocate based on seniority. Changing this expectation and proving that your time can be spent with greater use maybe worth it and is easier than fighting someone's worldview, but still a potential challenge.

PI might claim authorship even if the work is yours from cover to cover. You won't be able to protest, because you have to finish your PhD with them, then get recommendation letters etc. You may hope that they mention your leading role in the recommendation letters, but you will never know. Your claims that the idea and all the work is yours will be taken with a grain of salt: read other answers again and infer the consensus about the probability of a PhD student of 6 months coming up with a good idea. It could be the case that you generate and implement many good ideas by the time you finish (which will make your claims plausible), but it could be just one, this one.

Why would a PI do that? Don't they have myriads of their own better ideas? Well... Some don't. There are few practical upsides to being generous (and letting students take full credit for their work), as there is little danger in cutting little corners here and there. The first instinct of some high-score participants here is not to condemn PIs' unethical actions, but to highlight the importance of "having good relations with" a PI for one's own career (see the links in the 2nd paragraph).

Working on your own idea all by yourself can be a great motivating factor. You can be simply more efficient when you don't have to deal with the resistance or reluctance of others. Depending on your personality, it can be much easier to recover from wasting several weekends on a crappy idea ("embracing failure") than from the feeling of being dismissed, taken advantage of, or not given a proper credit. The latter can be more discouraging and destructive.

Previous answers pointed out that ideas expire, get published by others, are superseded by better ideas, and tend to get less exciting for you personally. Also, you will have less time and freedom as a postdoc. I agree that the optimal time for your best idea is now, the question is how. In addition to working on your idea with your PI and working on it alone, consider a third option: use your idea to establish a collaboration outside your PI's lab or your university:

  • Outside connections are good for your career: postdoc opportunities with that person, cross-fertilisation of research topics, independent feedback on your knowledge and skills, and a solid recommendation letter.
  • You PI might welcome this collaboration, provided there is no direct competition.
  • Even if your PI is included as a middle author, you'll find it easier to assert your primary role in the project, while others will find it easier to believe.
  • 1
    Thanks for your answer. This is an interesting perspective.
    – Time4Tea
    Mar 19, 2021 at 18:00

Keep in mind: given you are similarly bright-minded as your peers (including current PostDocs and Professors) and you are accessing the same material they are, it is likely that someone else is already working on those ideas.

So ... do not hold back them! On one hand, they can be the starting point of a path to your future career steps and those ideas can be your way out of a possible non-optimal development of your PhD: what if you feel strongly your ideas are correct, you can prove that, but your current advisor is strongly against those ideas? he may be right, but you better found out on your own if he is right or wrong... and you will need support from people in your field already investigating those ideas.

On the other hand, if you are just thinking about new possible numerical simulations or experiments that could be done, only to improve (not changing) current knowledge or getting more precise data, just piggy-back who is doing that already, experiments are something like 10% idea and 90% implementation, so proposing a new thing is worthwhile only if you can commit

  • your time or
  • some undergraduates' time working for you

Keep up your curiosity!


No matter how many ideas you have, they will only award you one PhD. The definition in most disciplines is an original contribution of knowledge (that is, your knowledge and not anyone else's). Hence, if you are sure of your facts you could certainly hold back ideas over and above "enough" if you have a good reason to do so. Many people, however, complete a PhD for personal reasons and in that case doing one's best would be the key driver.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .